> Would be interesting to train a model on images with alpha channels Would be even more interesting to get an ANN middle system of ontology of the (finally) r

> Would be interesting to train a model on images with alpha channels Would be even more interesting to get an ANN middle system of ontology of the (finally) represented content in order to change the single items. An internal representation of qualified structured items in space as part of the chain. Prompt > accessible internal representation > render.